Databases
StackOverflow2013
Postsdbo
PO
 

Note that there are some explanatory texts on larger screens.

plurals
  1. PO
    primarykey
    data
    text
    <p>You can bend over backwards trying to achieve this in pure SQL or you can create something a bit lengthier but both much easier to understand and more performance efficient - use a <a href="http://download.oracle.com/docs/cd/B19306_01/appdev.102/b14261/tuning.htm#LNPLS01210" rel="nofollow noreferrer">pipelined function</a></p> <p>In essence the function would accept a ref cursor which would have to be pre-ordered by the ID, and then pipe rows only when a contiguous block of records has ended.</p> <pre><code>CREATE TABLE assignment ( a_id NUMBER, assigned_to VARCHAR2(4000) ); CREATE OR REPLACE package PCK_CONTIGUOUS_GROUPBY as TYPE refcur_t IS REF CURSOR RETURN assignment%ROWTYPE; TYPE outrec_typ IS RECORD ( from_id NUMBER, to_id NUMBER, assigned_to VARCHAR2(4000)); TYPE outrecset IS TABLE OF outrec_typ; FUNCTION f_cont_groupby(p refcur_t) RETURN outrecset PIPELINED; end; / CREATE OR REPLACE package body pck_contiguous_groupby as FUNCTION f_cont_groupby(p refcur_t) RETURN outrecset PIPELINED IS out_rec outrec_typ; in_rec p%ROWTYPE; first_id assignment.a_id%type; last_id assignment.a_id%type; last_assigned_to assignment.assigned_to%type; BEGIN LOOP FETCH p INTO in_rec; EXIT WHEN p%NOTFOUND; IF last_id IS NULL THEN -- First record: don't pipe first_id := in_rec.a_id; ELSIF last_id = in_rec.a_id - 1 AND last_assigned_to = in_rec.assigned_to THEN -- Contiguous block: don't pipe NULL; ELSE -- Block not contiguous: pipe out_rec.from_id := first_id; out_rec.to_id := last_id; out_rec.assigned_to := last_assigned_to; PIPE ROW(out_rec); first_id := in_rec.a_id; END IF; last_id := in_rec.a_id; last_assigned_to := in_rec.assigned_to; END LOOP; CLOSE p; -- Pipe remaining row out_rec.from_id := first_id; out_rec.to_id := last_id; out_rec.assigned_to := last_assigned_to; PIPE ROW(out_rec); RETURN; END; END pck_contiguous_groupby; / </code></pre> <p>and then to try it out, populate the table and run:</p> <pre><code>SELECT * FROM TABLE(pck_contiguous_groupby.f_cont_groupby (CURSOR (SELECT a_id, assigned_to FROM assignment ORDER BY a_id))); </code></pre>
    singulars
    1. This table or related slice is empty.
    1. POOracle SQL for continuous grouping
      singulars
      1. PTQuestion
    1. PTAnswer
    1. This table or related slice is empty.
    1. USAndrew not the Saint
    plurals
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. This table or related slice is empty.
    1. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    2. VO
      singulars
      1. PO
      1. This table or related slice is empty.
      1. VTUpMod
    1. COPlsql row by row solution may be sometimes easier to read, but almost always is less performant than pure SQL.
      singulars
      1. PO
      1. USMichal Pravda
    2. COI'm also very curious why you state "and more performance efficient - use a pipelined function". Care to explain?
      singulars
      1. PO
      1. USRob van Wijk
    3. COIt is much more efficient than any other other SQL and I'll explain why. The first and foremost step of any performance design is not to place SQL ahead of PL/SQL but to minimize logical I/O. The above query performs a SINGLE SCAN of the table + an order by and that's it. Any other query would at least have to perform 2 scans. E.g. there probably is some clever way to use an analytic function to calculate the 'contiguous block' and then GROUP BY this new calculated column, but this would imply 2 scans + an additional GROUP BY.
      singulars
      1. PO
      1. USAndrew not the Saint
 

Querying!

 
Guidance
A row detail

Detail views are divided into sections. All the information in the data section comes from columns in the selected row. The other sections display data from other, related rows.

Related data can be related in a to-one or a to-many fashion. Captions of data related in a to-many fashion link to a list view showing a filtered view of the table.

Try moving around until you find a non-empty to-many entry and click on the label to get to one. You can move back to the root by clicking on the database name in the header.

SQuiL has stopped working due to an internal error.

If you are curious you may find further information in the browser console, which is accessible through the devtools (F12).

Reload